Understanding consumer engagement with vertical farming brands on Twitter: An approach of guided Latent Dirichlet Allocation for topic modeling

Shao, Junqi

Permalink

https://hdl.handle.net/2142/121499 Copy

Description

Title

Understanding consumer engagement with vertical farming brands on Twitter: An approach of guided Latent Dirichlet Allocation for topic modeling

Author(s)

Shao, Junqi

Issue Date

2023-07-21

Director of Research (if dissertation) or Advisor (if thesis)

Su, Leona Yi-Fan

Department of Study

Advertising

Discipline

Advertising

Degree Granting Institution

University of Illinois at Urbana-Champaign

Degree Name

M.S.

Degree Level

Thesis

Keyword(s)

• vertical farming
• consumer engagement
• startup communication
• content strategy
• sustainability
• topic modeling

Abstract

Vertical farming (VF) has emerged as a promising alternative to traditional farming practices in recent years. Despite its superiority in various aspects, it has been facing challenges in gaining enough consumer attention and acceptance. The present study first explores VF brand content strategies on social media, specifically on Twitter. The study attempts to identify diverse topics contained in VF brand tweets and classify these topics using a content typology that combines previous categorization frameworks with a data-driven approach. The study further examines how consumers engage with different types of VF brand content, manifested through the volume of “likes”, retweets, and replies. The guided Latent Dirichlet Allocation (LDA) was conducted to infer topics and themes underlying captured tweets from 12 popular VF brands. Negative binomial regression was also employed to predict consumer engagement with various brand content. The results indicate that VF brands primarily shared educational content, functional content, and relational content on Twitter while giving comparatively less prominence to remunerative content and corporate social responsibility (CSR) content. Nevertheless, it was observed that compared to tweets highlighting educational content, consumers are significantly less likely to engage with those highlighting functional content in terms of “likes”, retweets, and replies, while they are significantly more likely to engage with those focusing on CSR content in terms of all three metrics. Additionally, remunerative content and relational content were both significantly more likely to be “liked” and retweeted relative to educational content. These findings are expected to contribute to a theoretical understanding of how consumers engage with various social media content from novel food startups, as well as offer valuable insights for VF practitioners regarding the promotion of consumer acceptance toward the VF concept.

Graduation Semester

2023-08

Type of Resource

Thesis

Copyright and License Information

Copyright 2023 Junqi Shao

Owning Collections